METHOD AND APPARATUS FOR DE-DUPLICATION FOR SOLID STATE DISKs (SSDs)

1. A method of de-duplication comprising: receiving host data from a host, the host data being accompanied by a received host logical block number (LBA) and to be stored in solid state disks (SSDs); determining whether or not the received host data is the same as an existing host data, the existing host data being identifiable by an existing host LBA; upon determining that the received host data is the same as the existing host data, indexing a first table using the existing host LBA to retrieve an intermediate LBA (iLBA); linking the iLBA with the received host LBA in the first table, to avoid a write operation to the SSDs; upon determining that the received host data is not the same as the existing host data, identifying an available iLBA; linking the received host LBA to the identified iLBA; and writing the received host data to a location within the SSDs identified by use of the iLBA.

2. The method of de-duplication, as recited in claim 1 , wherein during the writing, identifying an available SSD LBA (SLBA) and saving the SLBA to a second table indexed by the identified iLBA so as to create an association between the identified iLBA and the available SLBA.

3. The method of deduplication, as recited in claim 1 , further including maintaining a count of the number of duplications of the host data.

4. The method of deduplication, as recited in claim 3 , wherein the count is incremented upon the determining that the received host data is the same as the existing host data.

5. The method of deduplication, as recited in claim 3 , wherein the count is decremented upon the step of determining that the received host data is not the same as the existing host data.

6. The method of deduplication, as recited in claim 1 , further including using an offset to address the SSDs when compressing the host data.

7. The method of deduplication, as recited in claim 1 , further including using meta data identifying host LBAs.

8. The method of deduplication, as recited in claim 7 , wherein during the writing, identifying an available SSD LBA (SLBA) and saving the SLBA to a second table indexed by the identified iLBA.

9. The method of deduplication, as recited in claim 8 , further including identifying valid SLBAs using the first and second tables.

10. The method of deduplication, as recited in claim 1 , further including identifying old SSD LBAs (SLBAs) and upon the old SLBAs matching the SLBAs of a stripe, performing logical garbage collection.

11. The method of deduplication, as recited in claim 1 , further including receiving a logical unit number (LUN) from the host and allocating a distinct first table to the LUN.

12. A method of de-duplication employing solid state disks (SSDs) comprising: receiving host data from a host, the host data being accompanied by a current host logical block number (LBA); upon the host data not being a duplicate host data, identifying an available intermediate LBA (iLBA); linking the current host LBA to the identified iLBA and during writing of the current host data, identifying an available SLBA and saving the SLBA to a second table indexed by the identified iLBA; writing the host data to the SSDs at a location identified by the SLBA; and upon subsequently receiving a duplicate of the host data accompanied by a different host LBA, linking the different host LBA to the identified iLBA and avoiding an actual write of the duplicate host data to the SSDs.

13. The method of de-duplication, as recited in claim 12 , wherein during the subsequently receiving step, finding a match in the SSDs between the current host data and the duplicate host data after receiving the duplicate host data and saving the identified iLBA to a location in a first table, indexed by the different host LBA so that the current host LBA and the different host LBA point to the identified iLBA in the first table.

14. The method of deduplication, as recited in claim 12 , further including maintaining a count of the number of duplications of the host data.

15. The method of deduplication, as recited in 12 , further including using an offset to address the SSDs when compressing the current host data and the different host data.

16. The method of deduplication, as recited in 12 , further including using meta data, identifying the host LBA.

17. The method of deduplication, as recited in claim 12 , further including upon old SLBAs matching the SLBAs of a stripe, performing logical garbage collection.

18. The method of deduplication, as recited in 12 , further including receiving a logical unit number (LUN) from the host, and allocating a distinct first table to the LUN, the distinct first table being one of the first tables.

19. A method of de-duplication comprising: receiving host data from a host, the received host data being accompanied by a received host logical block number (LBA) and to be stored in solid state disks (SSDs); determining whether or not the received host data is the same as an existing host data, the existing host data having linked therewith an intermediate (iLBA) and being identifiable by an existing host LBA; upon determining that the received host data is the same as the existing host data, linking the received host LBA with the iLBA to avoid a write operation to the SSDs; upon determining that the received host data is not the same as the existing host data, identifying an available iLBA; linking the received host LBA to the identified iLBA; and writing the received host data to a location within the SSDs identifiable by the iLBA.

20. The method of deduplication of claim 19 , wherein further receiving a logical unit number (LUN) from the host and generating a distinct first table associated with the LUN.